Journal of Medical Engineering and Technology最新文献

Retinal vessel segmentation is essential for precise ophthalmological diagnoses, particularly in the prediction of retinal degenerative diseases. However, existing methods usually lack robustness and accuracy, especially in segmentation of thin or overlapping vessels. To face these challenges, this study introduces an enhanced retina-RV-Gain segmentation model, which employs an architecture of various stages to refine the results of segmentation iteratively. The model integrates attention mechanisms to better capture complex vessel structures and employs an adaptive loss function to manage class imbalance. In addition, a specially designed discriminator enhances the model's ability to distinguish fine details from background noise vessels. The proposed RV-Gan is trained in comprehensive data sets that comprise retinal images, segmentation masks and noted labels, including Stare-DB, Chase-DB1 and Drive, using the Python platform. Experimental results demonstrate a segmentation accuracy of up to 99% in normal, abnormal and base vessels. These findings highlight the potential of the model to significantly improve diagnostic accuracy and support early prediction of disease in clinical ophthalmology. Overall, the enhanced RV-Gan architecture offers a robust solution to the limitations of current approaches, providing segmentation of high fidelity retinal vessels and advancing the predictive analysis of retinal degenerative conditions.

Sleep-disordered breathing (SDB) affects 14% of the population. Positive airway pressure (PAP) therapy is standard, but commercially available interfaces may be ineffective due to poor fit. Three-dimensional (3D) printing can customise PAP therapy interfaces. Is it feasible to manufacture and use 3D-printed customised oronasal PAP interfaces in clinical practice? Do customised interfaces improve patient comfort and reduce side effects compared to off-the-shelf interfaces? A single-site feasibility study involving 10 healthy and 10 patient participants was undertaken. A 3D facial scan was used to 3D print a mould, injected with medical-grade silicone to create a oronasal customised interface. Participants underwent a 10-minute trial with both off-the-shelf and customised interfaces. Comfort (Visual Analogue Scale), skin reactions, and interface leak (L/min) were measured. Patient participants used the customised interface for five nights at home, with data collected on Apnoea Hypopnoea Index (AHI), interface leak, and PAP therapy concordance. The study recruited 20 participants. Customised oronasal interfaces showed a failure rate in manufacturing (23.75% 3D printing, 50%: silicone injection). Adverse reactions were 10% in the patient study. Comfort scores were similar between interfaces. Interface leak was lower with customised interfaces after five nights. AHI was reduced with customised interfaces, but with a trend towards decreased PAP therapy concordance. The study demonstrated 3D-printed customised oronasal PAP interfaces can be manufactured, with potential benefits of reduced interface leak and AHI. Improvements in manufacturing processes are needed to reduce failure rates. Further research via a randomised controlled trial with a longer duration is warranted.

Deep learning's swift development has generated substantial excitement about its application in medical imaging. Machine learning (ML) methods can support radiologists in diagnosing breast cancer (BC) without resorting to invasive procedures. However, traditional ML classifiers require the extraction of detailed hand-crafted features, which is a time-intensive task to achieve accurate results. Hence, this paper proposes a novel Feature-driven Breast Cancer Classification using the Modified Loss and Activation function-assisted LeNet (MLAL) model, named F-BCC-ML. The process of detecting BC using mammogram images comprises several key stages. In the first step, the image undergoes enhancement using the Improved Bilateral Filtering Technique (IBFT), which reduces the noise while conserving critical structural details like edges. Next, the image is subjected to segmentation using SegNet, a deep-learning model designed for semantic segmentation. After segmentation, the next phase is feature extraction, where various features like Weber Local descriptor assisted Local Gabor XOR Pattern (WLD-LGXP) for texture analysis, Median Binary Pattern (MBP), colour features, and deep features are derived from the segmented image. Once the features are extracted, they are fed into the classification stage, where the Modified Loss and Activation function assisted LeNet (MLAL) model, more sophisticated Deep Convolutional Neural Network (DCNN) are used to classify the image as either normal or cancerous. The result is a prediction that indicates whether the breast tissue is benign or shows signs of cancer, helping radiologists make more accurate and informed decisions. The MLAL+DCNN accomplished the maximum accuracy of 0.936, precision of 0.947 and F-measure of 0.942, respectively.

Transcatheter aortic valve implantation (TAVI) is now the standard treatment for aortic stenosis, offering a less invasive alternative to surgery. While 3D printing and finite element analysis (FEA) show promise for pre-procedural planning, their accuracy in predicting post-TAVI device geometry remains unclear. This study evaluates the agreement between patient-specific FEA models, 3D-printed phantoms, and post-TAVI CT imaging in replicating implanted device geometry. Ten patients treated with the SAPIEN 3 Ultra (S3) device were analysed using pre- and post-TAVI CT scans. Both FEA simulations and 3D-printed models were assessed for stent deformation and anatomical fit. Agreement was quantified using statistical tools including concordance correlation coefficient (CCC), intraclass correlation coefficient (ICC), and Bland-Altman plots. FEA showed stronger agreement with post-TAVI CT (ICC = 0.614, CCC = 0.479) than 3D printing (ICC = 0.363, CCC = 0.165), which had higher variability. While FEA closely approximated device expansion at the annular level, both methods had limitations due to material and computational assumptions. The study supports the greater reliability of FEA in pre-procedural planning, highlighting the need for further validation and standardisation.

Capturing limb shape for amputees is critical in the fabrication and delivery of comfortable prosthetic limbs. Smartphone Photogrammetry offers a cheaper and more accessible alternative to digital shape capture than traditional handheld 3D scanners, opening possibilities for remote, or in home scanning. In this study we aimed to evaluate the accuracy of smartphone photogrammetry using a technique designed for in home scanning, comparing performance to an Einscan H2. The results indicated that photogrammetry was suitable accurate for scanning static limb targets (>95% volumetric accuracy), but was not accurate enough for direct amputee scanning (63.4% larger volumes). Whilst this technique was not sufficiently accurate for clinical use, the amputee surrogate trials did show increased accuracy, indicating the method shows promise and should be developed further, with a particular focus on home environment compatible techniques.

This study aims to highlight the noteworthy impression of the needle-free drug delivery devices to endorse drug delivery technology innovation. By briefing existing information, this assessment can guide the development of a new device. A thorough literature survey has been done to analyse the design, technology mechanism, CFD studies, clinical results, and patents filed in the field of such devices. Challenges and future scope of improvement in the existing devices were reported. A number of drug delivery devices were investigated and have been reported in this study. Among all the reported devices, the shock wave-operated device has the ability to reduce the current limitations in needle-free drug delivery device, offering a usable solution for treating diseases. Most devices were developed for liquid vaccination, and trials were done both on animals and humans. Clinical trial evidence shows that these systems were acceptable to clinicians as well as patients. Several parameters can be modified to attain the required depth of penetration under the skin.

A recent innovation in bioelectronic medicine is the use of implantable devices capable of harvesting biomechanical energy from cardiac motion. Such self-powered devices would facilitate cardiovascular functionality in patients with compromised hearts. This not only requires integrating bioelectronic medicine with cardiovascular physiology, but also a quantitative predictability of their functioning. We present a first attempt to establish a quantitative basis derived through biophysical considerations. Assuming cardiac functionality to be described using a spring-dashpot model, we present analytical solutions for different scenarios of physiological relevance. A key result is that the inverse lifetime lower than the natural frequency of the heart vibration leads to a rapid decrease in vibrational amplitudes of the implant as the cardiac cycle moves to the relaxation phase. When the inverse lifetime equals the natural frequency, vibrations persist to the largest extent and a substantial amount of energy can be harvested in a cardiac cycle via energy harvesting mechanisms (piezoelectric and triboelectric). Our analysis points to the critical role of the implant mass on variations in displacement during heart vibrations. Our theoretical predictions provide guidelines for developing next-generation biomedical devices with the heart as the in vivo source of energy harvesting.

A new artificial oesophagus is described. The device allows minimally invasive oesophageal resection and reconstruction with a new technology. It might permit the patient to live a life without the well-known restrictions after a gastric pull-up. The main functionality is an artificial muscle that continuously and actively transports the food, and a double acting reed valve that prohibits gastro-neo-oesophageal reflux, but allows vomiting and gas bloating. This device aims to bridge critical gaps in the field of oesophageal reconstruction using advanced mechanical systems.

Objective: Diabetic Peripheral Neuropathy (DPN) is the most common prolonged complication of diabetes. A nerve reaches to the hands, legs, and a foot is damaged due to excessive glucose level. This leads to the loss of sensation, numbness and pain in the feet, legs or hands. Currently available devices are expensive, take more time and need more expertise to operate them to detect the level of DPN. This study is designed to detect the level of diabetic peripheral neuropathy (DPN) from first joint of index finger using a novel 128-Hz electronic tuning fork prototype which is capable of performing accurate vibration perception duration (VPD).

Methods: A total of 169 diabetic patients were recruited from the secondary author's practice for assessment of level of DPN with our device. All the patients were enrolled according to an approved protocol. Patient places index finger on the tip of our device in such a way that the tip covers the first joint of index finger. Our device then provides the vibration of desired frequency and voltage to the index finger via tactile platform and patient starts feeling the vibration. Depending on the vibration perception duration (VPD) for which the patient feels the vibration, 4 levels of DPN i.e. Normal, Mild, Moderate and Severe are calculated. Three repeated measurements were taken from all 169 patients.

Results: Our device detected 74 DPN patients (6 severe, 26 moderates, 42 mild) and 89 normal (no DPN) patients. The mean of vibration perception duration (VPD) was 6.8 s, with a standard deviation (SD) of ± 0.84 s of all 169 patients. Mean VPD of severe, moderate, mild and normal level of DPN patients was 1.73 (mean SD = 0.7 s), 5.82 (mean SD = 0.84 s), 8.32 (mean SD = 1 s) and 11.3 s (mean SD = 0.84 s), respectively. Considering the Biothesiometer as the reference standard, our results were compared against it and our device's result accuracy was > 92%.

Conclusion: VPD was a sensitive measure of a detection of level of DPN. The device is compact, handy, easy to use and takes only few seconds to diagnose the level of DPN level in diabetic patients.